REVIEW ARTICLE Paul G. Barash, MD Giovanni Landoni, MD Section Editors Noninvasive Cardiac Output Monitors: A State-of the-Art Review Paul E. Marik, MD, FCCM, FCCP D ESPITE IMPROVEMENTS in resuscitation and support- ive care, progressive organ dysfunction occurs in a large proportion of patients with acute, life-threatening illnesses and those undergoing major surgery. 1-5 Recent data suggest that early aggressive resuscitation of critically ill patients may limit and/or reverse tissue hypoxia and progression to organ failure and improve outcome. 6 In a landmark study, Rivers et al 7 showed that a protocol of early goal-directed therapy reduces organ failure and improves survival in patients with severe sepsis and septic shock. Similarly, optimization of cardiac output (CO) in patients undergoing major surgery has been shown to reduce postoperative complications and the length of stay. 8-13 By contrast, excessive fluid resuscitation has been associated with increased complications, increased lengths of intensive care unit and hospital stay, and increased mortality. 14-17 These data suggest that fluid resuscitation should be titrated closely to minimize the risks of over- or under-resuscitation. 18 Over the last 2 decades, the understanding of the complex- ities of shock has improved, and conventional approaches to resuscitation have come under increasing scrutiny. The tradi- tional measured variables of resuscitation have included blood pressure, pulse rate, central venous pressure, and arterial oxy- gen saturation. These variables change minimally in early shock and are poor indicators of the adequacy of resuscita- tion. 19 Furthermore, the clinical assessment of CO and intra- vascular volume status are notoriously inaccurate. 20 With the increased recognition of the limitations of traditional methods to guide resuscitation, newer techniques have emerged that dynamically assess patients’ physiologic response to a hemo- dynamic challenge. In patients with indices of inadequate tissue perfusion, fluid resuscitation generally is regarded as the first step in resuscitation. However, clinical studies consistently have shown that only about 50% of hemodynamically unstable patients are volume responsive. 14 Therefore, the resuscita- tion of hemodynamically unstable patients requires an accu- rate assessment of the patients’ intravascular volume status (cardiac preload) and the ability to predict the hemodynamic response after a fluid challenge (volume responsiveness). Fundamentally, the only reason to give a patient a fluid challenge is to increase the stroke volume (SV) (volume responsiveness). If the fluid challenge does not increase the SV, volume loading serves the patient no useful benefit and is likely to be harmful. Therefore, the measurements of SV and CO are fundamental to the hemodynamic management of critically ill and injured patients and unstable patients in the operating room. Both fluid challenges and the use of inotropic agents/vasopressors should be based on the re- sponse of the SV to either of these challenges. Until re- cently, continuous real-time CO monitoring required a ther- modilution pulmonary artery catheter (PAC). During the past decade, several less invasive methods have been devel- oped. These technologies are reviewed in this article. Adolph Fick described the first method of CO estimation in 1870. 21 This method was the reference standard by which all other methods of determining CO were evaluated until the introduction of the PAC in the 1970s. 22 Despite its limitations, CO measurement with a PAC using the bolus thermodilution method has become the de facto gold standard for the measure- ment of CO and is the reference standard used to compare noninvasive technologies. 23,24 When assessing the reliability and clinical use of a noninvasive CO device, 2 factors are important: the accuracy of individual measurements compared with the reference standard and the ability to track changes in the SV and CO accurately and reproducibly after a therapeutic intervention. The latter is the most important factor when evaluating these devices because it directly impacts clinical decision making and therapeutic interventions. In most clinical situations, whether a cardiac index is 2.1 or 2.6 L/min/m 2 is not of great clinical importance; however, whether the change in the SV after a fluid bolus is 5% or 15% is of great clinical significance. The most frequently used analytic method for evaluating CO monitoring devices is the Bland-Altman method of plotting the bias against the mean CO and determining the limits of agreement (LOAs). 25 The percentage error is calcu- lated as the ratio of 2 standard deviations (SDs) of the bias (LOA) to the mean CO and is considered clinically acceptable if it is below 30%, as proposed by Critchley and Critchley. 23 The Bland-Altman method only addresses how well the method From the Division of Pulmonary and Critical Care Medicine, East- ern Virginia Medical School, Norfolk, VA. Address reprint requests to Paul E. Marik, MD, FCCM, FCCP, Eastern Virginia Medical School, 825 Fairfax Avenue, Suite 410, Norfolk, VA 23507. E-mail: [email protected] © 2012 Elsevier Inc. All rights reserved. 1053-0770/xx0x-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.03.022 Key words: hemodynamic monitoring, carbon dioxide rebreathing, bioimpedance, bioreactance, esophageal Doppler, pulse contour anal- ysis, cardiac output, stroke volume, goal-directed therapy, anesthesia, intensive care units, critical illness 1 Journal of Cardiothoracic and Vascular Anesthesia, Vol xx, No x (Month), 2012: pp xxx
Noninvasive Cardiac Output Monitors: A State-of the-Art Review
Feb 09, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
t r ( r F c r S i a o
J
Giovanni Landoni, MD Section Editors
Noninvasive Cardiac Output Monitors: A State-of the-Art Review
Paul E. Marik, MD, FCCM, FCCP
t i s c m p o
1 o i C m m n a i w t i e d s
DESPITE IMPROVEMENTS in resuscitation and support- ive care, progressive organ dysfunction occurs in a large
proportion of patients with acute, life-threatening illnesses and those undergoing major surgery.1-5 Recent data suggest that early aggressive resuscitation of critically ill patients may limit and/or reverse tissue hypoxia and progression to organ failure and improve outcome.6 In a landmark study, Rivers et al7
showed that a protocol of early goal-directed therapy reduces organ failure and improves survival in patients with severe sepsis and septic shock. Similarly, optimization of cardiac output (CO) in patients undergoing major surgery has been shown to reduce postoperative complications and the length of stay.8-13 By contrast, excessive fluid resuscitation has been associated with increased complications, increased lengths of intensive care unit and hospital stay, and increased mortality.14-17 These data suggest that fluid resuscitation should be titrated closely to minimize the risks of over- or under-resuscitation.18
Over the last 2 decades, the understanding of the complex- ities of shock has improved, and conventional approaches to resuscitation have come under increasing scrutiny. The tradi- tional measured variables of resuscitation have included blood pressure, pulse rate, central venous pressure, and arterial oxy- gen saturation. These variables change minimally in early shock and are poor indicators of the adequacy of resuscita- tion.19 Furthermore, the clinical assessment of CO and intra- vascular volume status are notoriously inaccurate.20 With the increased recognition of the limitations of traditional methods to guide resuscitation, newer techniques have emerged that dynamically assess patients’ physiologic response to a hemo- dynamic challenge.
In patients with indices of inadequate tissue perfusion, fluid resuscitation generally is regarded as the first step in resuscitation. However, clinical studies consistently have shown that only about 50% of hemodynamically unstable patients are volume responsive.14 Therefore, the resuscita- ion of hemodynamically unstable patients requires an accu- ate assessment of the patients’ intravascular volume status cardiac preload) and the ability to predict the hemodynamic esponse after a fluid challenge (volume responsiveness). undamentally, the only reason to give a patient a fluid hallenge is to increase the stroke volume (SV) (volume esponsiveness). If the fluid challenge does not increase the V, volume loading serves the patient no useful benefit and
s likely to be harmful. Therefore, the measurements of SV nd CO are fundamental to the hemodynamic management
f critically ill and injured patients and unstable patients in
ournal of Cardiothoracic and Vascular Anesthesia, Vol xx, No x (Month),
he operating room. Both fluid challenges and the use of notropic agents/vasopressors should be based on the re- ponse of the SV to either of these challenges. Until re- ently, continuous real-time CO monitoring required a ther- odilution pulmonary artery catheter (PAC). During the
ast decade, several less invasive methods have been devel- ped. These technologies are reviewed in this article. Adolph Fick described the first method of CO estimation in
870.21 This method was the reference standard by which all ther methods of determining CO were evaluated until the ntroduction of the PAC in the 1970s.22 Despite its limitations, O measurement with a PAC using the bolus thermodilution ethod has become the de facto gold standard for the measure- ent of CO and is the reference standard used to compare
oninvasive technologies.23,24 When assessing the reliability nd clinical use of a noninvasive CO device, 2 factors are mportant: the accuracy of individual measurements compared ith the reference standard and the ability to track changes in
he SV and CO accurately and reproducibly after a therapeutic ntervention. The latter is the most important factor when valuating these devices because it directly impacts clinical ecision making and therapeutic interventions. In most clinical ituations, whether a cardiac index is 2.1 or 2.6 L/min/m2 is not
of great clinical importance; however, whether the change in the SV after a fluid bolus is 5% or 15% is of great clinical significance. The most frequently used analytic method for evaluating CO monitoring devices is the Bland-Altman method of plotting the bias against the mean CO and determining the limits of agreement (LOAs).25 The percentage error is calcu- lated as the ratio of 2 standard deviations (SDs) of the bias (LOA) to the mean CO and is considered clinically acceptable if it is below 30%, as proposed by Critchley and Critchley.23
The Bland-Altman method only addresses how well the method
From the Division of Pulmonary and Critical Care Medicine, East- ern Virginia Medical School, Norfolk, VA.
Address reprint requests to Paul E. Marik, MD, FCCM, FCCP, Eastern Virginia Medical School, 825 Fairfax Avenue, Suite 410, Norfolk, VA 23507. E-mail: [email protected]
© 2012 Elsevier Inc. All rights reserved. 1053-0770/xx0x-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.03.022 Key words: hemodynamic monitoring, carbon dioxide rebreathing,
bioimpedance, bioreactance, esophageal Doppler, pulse contour anal- ysis, cardiac output, stroke volume, goal-directed therapy, anesthesia,
intensive care units, critical illness
12012: pp xxx
i
f a a v T c m L
2 PAUL E. MARIK
being evaluated agrees with the reference method and fails to show whether the test method reliably detects changes in CO. Although the accuracy of noninvasive CO devices to measure trends in CO has not been standardized, a number of methods have been described in the literature, including the correlation coefficient, the Bland-Altman method, the 4-quadrant plot, and receiver-operator characteristic (ROC) curve analysis.24
CO AS MEASURED BY CARBON DIOXIDE REBREATHING
CO can be calculated by the CO2 partial rebreathing tech- ique using the modified Fick equation.21 NICO (Respironics, urraysville, PA) is a proprietary device that measures CO
ased on this principle. The CO2 partial rebreathing technique ompares end-tidal carbon dioxide partial pressure obtained uring a nonrebreathing period with that obtained during a ubsequent rebreathing period. The ratio of the change in nd-tidal carbon dioxide and CO2 elimination after a brief
period of partial rebreathing (usually 50 seconds) provides a noninvasive estimate of the CO.26 A limitation of the rebreath- ng CO2 CO method is that it only measures pulmonary capil-
lary blood flow (ie, the nonshunted portion of the CO). To calculate the total CO, intrapulmonary shunt and anatomic shunt factions (Qs/Qt) must be added to the pulmonary capil- lary blood flow. The NICO system estimates Qs/Qt using a shunt correction algorithm that uses oxygen saturation from pulse oximetry and the fractional concentration of inspired oxygen.
The CO2 rebreathing technique has a number of significant limitations. Almost all the validation studies have been per- formed in patients undergoing anesthesia or in deeply sedated mechanically ventilated intensive care unit patients in whom the agreement with thermodilution CO has varied from “poor” to “acceptable.”27-32 In spontaneously breathing patients, the ebreathing period is associated with an increase in minute entilation.33 This reduces the accuracy of the CO determina- ions.30,34 Furthermore, a low minute ventilation, high shunt raction, and a high CO result in inaccurate measure- ents.27,29,34 Considering the limitations of this technology and
he potential inaccuracies, the routine use of the CO2 rebreath- ing technique to guide fluid and vasopressor therapy cannot be recommended.
ESOPHAGEAL DOPPLER
The esophageal Doppler technique measures blood flow ve- locity in the descending aorta by means of a Doppler transducer (4-MHz continuous wave or 5-MHz pulsed wave according to the manufacturers) placed at the tip of a flexible probe. The probe is introduced into the esophagus of sedated, mechanically ventilated patients and then rotated so the transducer faces the descending aorta and a characteristic aortic velocity signal is obtained. The CO is calculated based on the diameter of the aorta (measured or estimated), the distribution of the CO to the descending aorta, and the measured flow velocity of blood in the aorta. Because esophageal Doppler probes are inserted blindly, the resulting waveform is highly dependent on correct positioning. The clinician must adjust the depth, rotate the probe, and adjust the gain to obtain an optimal signal.35 Poor
positioning of the esophageal probe tends to underestimate the n
true CO. There is a significant learning curve in obtaining adequate Doppler signals, and the correlations are better in studies in which the investigator was not blinded to the results of the CO obtained with a PAC.36 A major limitation of esophageal Doppler monitoring is the assumption that a fixed percentage of the CO is directed to the head and descending aorta. Although this may be true in healthy volunteers, the present authors have shown that a disproportionate percentage of the increase in CO with fluid loading in hemodynamically unstable patients is directed into the carotid arteries.37 There- fore, the increase in blood flow velocity in the descending aorta may not correlate well with the increase in the SV. Neverthe- less, esophageal Doppler monitoring has use in aiding in the assessment of the hemodynamic status and guiding fluid ther- apy in the operating room.10-12,38
A completely noninvasive Doppler technology, the ultra- sound CO monitor (USCOM, Sydney, Australia), uses transaortic or transpulmonary Doppler ultrasound flow trac- ings to calculate CO as the product of the SV and heart rate. The SV is calculated from a proprietary algorithm applying ultrasound principles of blood velocity–time integral (VTI) measurements in the ventricular aortic/pulmonary outflow tract. Studies comparing USCOM measurements of CO with those obtained by the standard thermodilution technique have shown mixed results.39-42 The use of Doppler ultra- sound to determine the cardiac index has several inherent technologic limitations. Potential sources of variation exist in the estimation of the aortic/pulmonary outflow tract area, the determination of the VTI, and the variability with oper- ator-dependent measurements. With USCOM, the aortic/ pulmonary outflow tract area is not measured directly but rather calculated from a proprietary anthropometric algo- rithm based on the subject’s body height. The stroke dis- tance is simply the distance a red blood cell travels per systolic stroke. This is measured as the VTI of the Doppler flow profile of each systolic stroke. Thus, the accuracy of the USCOM technology depends on obtaining accurate, repro- ducible VTI values. A precise VTI measurement requires a good flow signal and its correct interpretation, both of which are heavily dependent on the subject and the operator. An improper technique of poor Doppler ultrasound beam align- ment with blood flow at the aortic/pulmonary outflow tract will lead to suboptimal VTI measurements. A further limi- tation of this technique is that it is not conducive to contin- uous monitoring.
PULSE CONTOUR ANALYSIS
The concept of pulse contour analysis is based on the relation among blood pressure, SV, arterial compliance, and systemic vascular resistance (SVR).43 The SV or CO can be calculated rom the arterial pressure waveform if the arterial compliance nd SVR are known. Although the 4 pulse contour systems that re available commercially use different pressure-volume con- ersion algorithms, they are based on this basic principle. hese systems can be divided into 3 categories: (1) pulse ontour analysis requiring an indicator dilution CO measure- ent to calibrate the pulse contour (ie, LiDCO System; iDCO, Cambridge, UK; and PiCCO System; Pulsion, Mu-
ich, Germany), (2) pulse contour analysis requiring patient
h v
T C
y d r a s c t t a c r
c
3NONINVASIVE CARDIAC OUTPUT MONITORS
demographic and physical characteristics for arterial imped- ance estimation (ie, FloTrac System; Edwards Lifesciences, Irvine, CA), and (3) pulse contour analysis that does not require calibration or preloaded data (ie, MostCare System; Vyetech Health, Padua, Italy). Table 1 contrasts the characteristics of the 4 systems. In addition to measuring the SV, these systems report the SV variation (SVV) and/or pulse pressure variation (PPV). The SVV/PPV may be useful in predicting fluid respon- siveness in select patient groups (see later).
An important factor when interpreting the CO measured by a pulse contour system is the site that the blood pressure is measured (ie, the radial v the femoral artery). Discrepancies among central and peripheral blood pressures have been de- scribed in a number of clinical circumstances, such as after cardiopulmonary bypass, in patients with septic shock treated with high-dose vasoconstrictors, and in patients during reper- fusion after a liver transplant.44 The differences in blood pres- sure among different sites may be large, and in conditions of intense vasoconstriction, the radial blood pressure may under- estimate the true aortic blood pressure, giving a falsely low CO value. Furthermore, it has been shown that in volume-respon- sive patients there is selective redistribution of blood flow to the cerebral circulation with a significantly smaller percentage increase in blood flow in the brachial artery.37 This may lead to a significant error when the radial pulse is used for pulse contour analysis.
Lithium Dilution and Pulse Contour Analysis
The LiDCO system combines pulse contour analysis with lithium indicator dilution for continuous SV and SVV moni- toring. The arterial pressure waveform is interpreted as a con- tinuous curve describing the volume of the arterial tree in arbitrary units (standardized volume waveform). The effective value (approximately 0.7 times the original amplitude) of this
Table 1. Overview of the Pulse Contou
System Characteristic FloTrac System PiCCO
Arterial waveform analysis
Area under the s the arterial wa
Requirements Peripheral or central arterial catheter
Central arterial ca subclavian or I
Calibration Uncalibrated/internal Transpulmonary
Operator independent Easy to use
Broad range of h parameters
More robust duri instability
More invasive
Adapted with permission.43
Abbreviations: CVC, central venous catheter; GEDV, global end-dia root mean square.
volume waveform is determined using the root mean square, a u
mathematic principle to calculate the magnitude of a varying quantity. The root mean square value is called “nominal SV” and is scaled to an “actual SV” using a patient-specific cali- bration factor.43 This factor is derived from a lithium indicator dilution CO measurement and corrects for arterial compliance and variations among individuals. The lithium can be injected into a peripheral vein, and the doses do not exert pharmaco- logically relevant effects in adult patients. The LiDCO indica- tor dilution method has shown to be at least as reliable as other thermodilution methods over a broad range of CO in a variety of patients.45-48 Recalibration should be performed after acute emodynamic changes and after any intervention that alters ascular impedance.
ranspulmonary Thermodilution and Pulse ontour Analysis
The PiCCO monitoring system combines pulse contour anal- sis with the transpulmonary thermodilution CO (TPCO) to etermine a number of hemodynamic parameters. The TPCO equires both central venous (internal jugular or subclavian) nd central arterial (femoral artery) catheterization. TPCO mea- urements with PiCCO have been shown to be reliable in omparison with PAC thermodilution in broad groups of pa- ients.49,50 The continuous pulse contour SV is calculated from he area under the systolic portion of the arterial waveform. In ddition, the shape of the arterial waveform (dP/dt), arterial ompliance, SVR, and a patient-specific calibration factor are equired for the calculation.43,49 Arterial compliance is derived
from the SVR and the shape of the diastolic part of the arterial waveform. The PiCCO monitor uses TPCO measurement for calibration of the algorithm. The PiCCO calibration appears to remain accurate within 6 hours of calibration even when the vascular tone has changed.51 In addition, the thermodilution urve can be used to measure the global end-diastolic vol-
d Hemodynamic Monitoring Devices43
LiDCO System PRAM
c portion of RMS method applied to the arterial pressure signal
Area under curve
Peripheral or central arterial catheter
odilution Lithium indicator dilution Uncalibrated/internal Manual Automatic Manual Lithium None SVV, PPV, GEDV, EVLW, SVR SVV, PPV
ynamic
modynamic
hemodynamic instability
Minimally invasive
volume; EVLW, extravascular lung water; IJ, internal jugular; RMS,
r-Base
System
b
P P
c p e t S
a s p d w a w a o i r c
g H s v n o
h
P N
t
4 PAUL E. MARIK
edema.49,52-54 The monitor also measures SVV/PVV, which has een shown to be predictive of fluid responsiveness.55 In a
randomized controlled trial, Mutoh et al56 showed an improved linical outcome for patients with subarachnoid hemorrhage andomized to a PiCCO-based hemodynamic algorithm as ompared with the “standard of care,” which used a PAC lgorithm. Additional studies are required to evaluate the clin- cal benefit of this technology.
ulse Contour Requiring Patient Demographic and hysical Characteristics and No Calibration
The FloTrac system consists of the FloTrac sensor and orresponding Vigileo monitor. The system is operator inde- endent, needs no external calibration, and requires a periph- ral arterial catheter only. The basic principle of the system is he linear relation between the pulse pressure and the SV. The V is estimated using the following equation43: SV SDAP
. The arterial pressure waveform is sampled each 20 seconds t 100 Hz, which results in 2,000 data points. SDAP is the tandard deviation of these data points and reflects the pulse ressure. The factor represents the conversion factor that epends on arterial compliance, the mean arterial pressure, and aveform characteristics. The patient’s vascular compliance is
ssessed using biometric values (ie, sex, age, height, and eight) according to the method described by Langewouters et
l.57 Waveform characteristics assessed are skewness (degree f asymmetry) and kurtosis (degree of peakedness) of the ndividual arterial pressure curve. Skewness and kurtosis rep- esent changes in the arterial waveform, which should reflect hanges in vascular tone. The factor is recalculated every
minute and enables calculation of the SV without external calibration.
Because the system is operator independent, easy to use, needs no external calibration, and only requires a peripheral arterial catheter (usually the radial artery), the FloTrac system has found popular appeal and has been studied widely, partic- ularly in the setting of cardiac surgery. To date, the accuracy of FloTrac has been evaluated in 45 studies58-102; these are sum- marized in Table 2. Studies evaluating the first-generation FloTrac showed poor agreement compared with intermittent thermodilution, which is the gold standard. Second-generation devices were purported to be more reliable; however, their accuracy remained clinically unacceptable. Furthermore, in pa- tients with low SVR (eg, sepsis or liver failure), measurements were unreliable, with the bias being correlated with the SVR.65,95,99,101 The data from Table 2 show that the percentage error is lower in cardiac patients as compared with other cohorts (37% 11% v 47% 11%, p 0.01). The third- eneration software claims to have overcome these problems. owever, 6 recent validation studies evaluating this latest ver-
ion do not show improved accuracy in comparison with older ersions.97-102 More problematic is the fact that the system does ot track changes in the SV accurately after a volume challenge r after the use of vasopressors.65,79,85,94,95,100,101,103 These lim-
itations significantly restrict the clinical use of this device. The SVV may be useful in intraoperative fluid optimization in select noncardiac surgical patients.104 However, in a co-
ort of medical patients, it was reported that the SVV was d
oorly predictive of volume responsiveness.105 Takala et l106 randomized 388 hemodynamically unstable patients to oninvasive monitoring with the FloTrac system for 24 ours or usual care (the control group). The main outcome easure was the proportion of patients achieving hemody-
amic stability within 6 hours of starting the study. Surpris- ngly, the time to reach the predefined resuscitation goals as longer in the FloTrac group, with worse clinical out-
omes in these patients.
ulse Contour Requiring No Patient Data and o Calibration
The MostCare system uses the pressure recording analytic ethod (PRAM) to determine the SV.107 PRAM measures the
area under the curve of the arterial waveform. No external calibration or pre-estimated data are required. The morphology of the arterial waveform is analyzed to determine an internal calibration. This system uses high-time resolution by sampling the signal at 1,000 Hz, and it analyzes the whole cardiac…
J
Giovanni Landoni, MD Section Editors
Noninvasive Cardiac Output Monitors: A State-of the-Art Review
Paul E. Marik, MD, FCCM, FCCP
t i s c m p o
1 o i C m m n a i w t i e d s
DESPITE IMPROVEMENTS in resuscitation and support- ive care, progressive organ dysfunction occurs in a large
proportion of patients with acute, life-threatening illnesses and those undergoing major surgery.1-5 Recent data suggest that early aggressive resuscitation of critically ill patients may limit and/or reverse tissue hypoxia and progression to organ failure and improve outcome.6 In a landmark study, Rivers et al7
showed that a protocol of early goal-directed therapy reduces organ failure and improves survival in patients with severe sepsis and septic shock. Similarly, optimization of cardiac output (CO) in patients undergoing major surgery has been shown to reduce postoperative complications and the length of stay.8-13 By contrast, excessive fluid resuscitation has been associated with increased complications, increased lengths of intensive care unit and hospital stay, and increased mortality.14-17 These data suggest that fluid resuscitation should be titrated closely to minimize the risks of over- or under-resuscitation.18
Over the last 2 decades, the understanding of the complex- ities of shock has improved, and conventional approaches to resuscitation have come under increasing scrutiny. The tradi- tional measured variables of resuscitation have included blood pressure, pulse rate, central venous pressure, and arterial oxy- gen saturation. These variables change minimally in early shock and are poor indicators of the adequacy of resuscita- tion.19 Furthermore, the clinical assessment of CO and intra- vascular volume status are notoriously inaccurate.20 With the increased recognition of the limitations of traditional methods to guide resuscitation, newer techniques have emerged that dynamically assess patients’ physiologic response to a hemo- dynamic challenge.
In patients with indices of inadequate tissue perfusion, fluid resuscitation generally is regarded as the first step in resuscitation. However, clinical studies consistently have shown that only about 50% of hemodynamically unstable patients are volume responsive.14 Therefore, the resuscita- ion of hemodynamically unstable patients requires an accu- ate assessment of the patients’ intravascular volume status cardiac preload) and the ability to predict the hemodynamic esponse after a fluid challenge (volume responsiveness). undamentally, the only reason to give a patient a fluid hallenge is to increase the stroke volume (SV) (volume esponsiveness). If the fluid challenge does not increase the V, volume loading serves the patient no useful benefit and
s likely to be harmful. Therefore, the measurements of SV nd CO are fundamental to the hemodynamic management
f critically ill and injured patients and unstable patients in
ournal of Cardiothoracic and Vascular Anesthesia, Vol xx, No x (Month),
he operating room. Both fluid challenges and the use of notropic agents/vasopressors should be based on the re- ponse of the SV to either of these challenges. Until re- ently, continuous real-time CO monitoring required a ther- odilution pulmonary artery catheter (PAC). During the
ast decade, several less invasive methods have been devel- ped. These technologies are reviewed in this article. Adolph Fick described the first method of CO estimation in
870.21 This method was the reference standard by which all ther methods of determining CO were evaluated until the ntroduction of the PAC in the 1970s.22 Despite its limitations, O measurement with a PAC using the bolus thermodilution ethod has become the de facto gold standard for the measure- ent of CO and is the reference standard used to compare
oninvasive technologies.23,24 When assessing the reliability nd clinical use of a noninvasive CO device, 2 factors are mportant: the accuracy of individual measurements compared ith the reference standard and the ability to track changes in
he SV and CO accurately and reproducibly after a therapeutic ntervention. The latter is the most important factor when valuating these devices because it directly impacts clinical ecision making and therapeutic interventions. In most clinical ituations, whether a cardiac index is 2.1 or 2.6 L/min/m2 is not
of great clinical importance; however, whether the change in the SV after a fluid bolus is 5% or 15% is of great clinical significance. The most frequently used analytic method for evaluating CO monitoring devices is the Bland-Altman method of plotting the bias against the mean CO and determining the limits of agreement (LOAs).25 The percentage error is calcu- lated as the ratio of 2 standard deviations (SDs) of the bias (LOA) to the mean CO and is considered clinically acceptable if it is below 30%, as proposed by Critchley and Critchley.23
The Bland-Altman method only addresses how well the method
From the Division of Pulmonary and Critical Care Medicine, East- ern Virginia Medical School, Norfolk, VA.
Address reprint requests to Paul E. Marik, MD, FCCM, FCCP, Eastern Virginia Medical School, 825 Fairfax Avenue, Suite 410, Norfolk, VA 23507. E-mail: [email protected]
© 2012 Elsevier Inc. All rights reserved. 1053-0770/xx0x-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.03.022 Key words: hemodynamic monitoring, carbon dioxide rebreathing,
bioimpedance, bioreactance, esophageal Doppler, pulse contour anal- ysis, cardiac output, stroke volume, goal-directed therapy, anesthesia,
intensive care units, critical illness
12012: pp xxx
i
f a a v T c m L
2 PAUL E. MARIK
being evaluated agrees with the reference method and fails to show whether the test method reliably detects changes in CO. Although the accuracy of noninvasive CO devices to measure trends in CO has not been standardized, a number of methods have been described in the literature, including the correlation coefficient, the Bland-Altman method, the 4-quadrant plot, and receiver-operator characteristic (ROC) curve analysis.24
CO AS MEASURED BY CARBON DIOXIDE REBREATHING
CO can be calculated by the CO2 partial rebreathing tech- ique using the modified Fick equation.21 NICO (Respironics, urraysville, PA) is a proprietary device that measures CO
ased on this principle. The CO2 partial rebreathing technique ompares end-tidal carbon dioxide partial pressure obtained uring a nonrebreathing period with that obtained during a ubsequent rebreathing period. The ratio of the change in nd-tidal carbon dioxide and CO2 elimination after a brief
period of partial rebreathing (usually 50 seconds) provides a noninvasive estimate of the CO.26 A limitation of the rebreath- ng CO2 CO method is that it only measures pulmonary capil-
lary blood flow (ie, the nonshunted portion of the CO). To calculate the total CO, intrapulmonary shunt and anatomic shunt factions (Qs/Qt) must be added to the pulmonary capil- lary blood flow. The NICO system estimates Qs/Qt using a shunt correction algorithm that uses oxygen saturation from pulse oximetry and the fractional concentration of inspired oxygen.
The CO2 rebreathing technique has a number of significant limitations. Almost all the validation studies have been per- formed in patients undergoing anesthesia or in deeply sedated mechanically ventilated intensive care unit patients in whom the agreement with thermodilution CO has varied from “poor” to “acceptable.”27-32 In spontaneously breathing patients, the ebreathing period is associated with an increase in minute entilation.33 This reduces the accuracy of the CO determina- ions.30,34 Furthermore, a low minute ventilation, high shunt raction, and a high CO result in inaccurate measure- ents.27,29,34 Considering the limitations of this technology and
he potential inaccuracies, the routine use of the CO2 rebreath- ing technique to guide fluid and vasopressor therapy cannot be recommended.
ESOPHAGEAL DOPPLER
The esophageal Doppler technique measures blood flow ve- locity in the descending aorta by means of a Doppler transducer (4-MHz continuous wave or 5-MHz pulsed wave according to the manufacturers) placed at the tip of a flexible probe. The probe is introduced into the esophagus of sedated, mechanically ventilated patients and then rotated so the transducer faces the descending aorta and a characteristic aortic velocity signal is obtained. The CO is calculated based on the diameter of the aorta (measured or estimated), the distribution of the CO to the descending aorta, and the measured flow velocity of blood in the aorta. Because esophageal Doppler probes are inserted blindly, the resulting waveform is highly dependent on correct positioning. The clinician must adjust the depth, rotate the probe, and adjust the gain to obtain an optimal signal.35 Poor
positioning of the esophageal probe tends to underestimate the n
true CO. There is a significant learning curve in obtaining adequate Doppler signals, and the correlations are better in studies in which the investigator was not blinded to the results of the CO obtained with a PAC.36 A major limitation of esophageal Doppler monitoring is the assumption that a fixed percentage of the CO is directed to the head and descending aorta. Although this may be true in healthy volunteers, the present authors have shown that a disproportionate percentage of the increase in CO with fluid loading in hemodynamically unstable patients is directed into the carotid arteries.37 There- fore, the increase in blood flow velocity in the descending aorta may not correlate well with the increase in the SV. Neverthe- less, esophageal Doppler monitoring has use in aiding in the assessment of the hemodynamic status and guiding fluid ther- apy in the operating room.10-12,38
A completely noninvasive Doppler technology, the ultra- sound CO monitor (USCOM, Sydney, Australia), uses transaortic or transpulmonary Doppler ultrasound flow trac- ings to calculate CO as the product of the SV and heart rate. The SV is calculated from a proprietary algorithm applying ultrasound principles of blood velocity–time integral (VTI) measurements in the ventricular aortic/pulmonary outflow tract. Studies comparing USCOM measurements of CO with those obtained by the standard thermodilution technique have shown mixed results.39-42 The use of Doppler ultra- sound to determine the cardiac index has several inherent technologic limitations. Potential sources of variation exist in the estimation of the aortic/pulmonary outflow tract area, the determination of the VTI, and the variability with oper- ator-dependent measurements. With USCOM, the aortic/ pulmonary outflow tract area is not measured directly but rather calculated from a proprietary anthropometric algo- rithm based on the subject’s body height. The stroke dis- tance is simply the distance a red blood cell travels per systolic stroke. This is measured as the VTI of the Doppler flow profile of each systolic stroke. Thus, the accuracy of the USCOM technology depends on obtaining accurate, repro- ducible VTI values. A precise VTI measurement requires a good flow signal and its correct interpretation, both of which are heavily dependent on the subject and the operator. An improper technique of poor Doppler ultrasound beam align- ment with blood flow at the aortic/pulmonary outflow tract will lead to suboptimal VTI measurements. A further limi- tation of this technique is that it is not conducive to contin- uous monitoring.
PULSE CONTOUR ANALYSIS
The concept of pulse contour analysis is based on the relation among blood pressure, SV, arterial compliance, and systemic vascular resistance (SVR).43 The SV or CO can be calculated rom the arterial pressure waveform if the arterial compliance nd SVR are known. Although the 4 pulse contour systems that re available commercially use different pressure-volume con- ersion algorithms, they are based on this basic principle. hese systems can be divided into 3 categories: (1) pulse ontour analysis requiring an indicator dilution CO measure- ent to calibrate the pulse contour (ie, LiDCO System; iDCO, Cambridge, UK; and PiCCO System; Pulsion, Mu-
ich, Germany), (2) pulse contour analysis requiring patient
h v
T C
y d r a s c t t a c r
c
3NONINVASIVE CARDIAC OUTPUT MONITORS
demographic and physical characteristics for arterial imped- ance estimation (ie, FloTrac System; Edwards Lifesciences, Irvine, CA), and (3) pulse contour analysis that does not require calibration or preloaded data (ie, MostCare System; Vyetech Health, Padua, Italy). Table 1 contrasts the characteristics of the 4 systems. In addition to measuring the SV, these systems report the SV variation (SVV) and/or pulse pressure variation (PPV). The SVV/PPV may be useful in predicting fluid respon- siveness in select patient groups (see later).
An important factor when interpreting the CO measured by a pulse contour system is the site that the blood pressure is measured (ie, the radial v the femoral artery). Discrepancies among central and peripheral blood pressures have been de- scribed in a number of clinical circumstances, such as after cardiopulmonary bypass, in patients with septic shock treated with high-dose vasoconstrictors, and in patients during reper- fusion after a liver transplant.44 The differences in blood pres- sure among different sites may be large, and in conditions of intense vasoconstriction, the radial blood pressure may under- estimate the true aortic blood pressure, giving a falsely low CO value. Furthermore, it has been shown that in volume-respon- sive patients there is selective redistribution of blood flow to the cerebral circulation with a significantly smaller percentage increase in blood flow in the brachial artery.37 This may lead to a significant error when the radial pulse is used for pulse contour analysis.
Lithium Dilution and Pulse Contour Analysis
The LiDCO system combines pulse contour analysis with lithium indicator dilution for continuous SV and SVV moni- toring. The arterial pressure waveform is interpreted as a con- tinuous curve describing the volume of the arterial tree in arbitrary units (standardized volume waveform). The effective value (approximately 0.7 times the original amplitude) of this
Table 1. Overview of the Pulse Contou
System Characteristic FloTrac System PiCCO
Arterial waveform analysis
Area under the s the arterial wa
Requirements Peripheral or central arterial catheter
Central arterial ca subclavian or I
Calibration Uncalibrated/internal Transpulmonary
Operator independent Easy to use
Broad range of h parameters
More robust duri instability
More invasive
Adapted with permission.43
Abbreviations: CVC, central venous catheter; GEDV, global end-dia root mean square.
volume waveform is determined using the root mean square, a u
mathematic principle to calculate the magnitude of a varying quantity. The root mean square value is called “nominal SV” and is scaled to an “actual SV” using a patient-specific cali- bration factor.43 This factor is derived from a lithium indicator dilution CO measurement and corrects for arterial compliance and variations among individuals. The lithium can be injected into a peripheral vein, and the doses do not exert pharmaco- logically relevant effects in adult patients. The LiDCO indica- tor dilution method has shown to be at least as reliable as other thermodilution methods over a broad range of CO in a variety of patients.45-48 Recalibration should be performed after acute emodynamic changes and after any intervention that alters ascular impedance.
ranspulmonary Thermodilution and Pulse ontour Analysis
The PiCCO monitoring system combines pulse contour anal- sis with the transpulmonary thermodilution CO (TPCO) to etermine a number of hemodynamic parameters. The TPCO equires both central venous (internal jugular or subclavian) nd central arterial (femoral artery) catheterization. TPCO mea- urements with PiCCO have been shown to be reliable in omparison with PAC thermodilution in broad groups of pa- ients.49,50 The continuous pulse contour SV is calculated from he area under the systolic portion of the arterial waveform. In ddition, the shape of the arterial waveform (dP/dt), arterial ompliance, SVR, and a patient-specific calibration factor are equired for the calculation.43,49 Arterial compliance is derived
from the SVR and the shape of the diastolic part of the arterial waveform. The PiCCO monitor uses TPCO measurement for calibration of the algorithm. The PiCCO calibration appears to remain accurate within 6 hours of calibration even when the vascular tone has changed.51 In addition, the thermodilution urve can be used to measure the global end-diastolic vol-
d Hemodynamic Monitoring Devices43
LiDCO System PRAM
c portion of RMS method applied to the arterial pressure signal
Area under curve
Peripheral or central arterial catheter
odilution Lithium indicator dilution Uncalibrated/internal Manual Automatic Manual Lithium None SVV, PPV, GEDV, EVLW, SVR SVV, PPV
ynamic
modynamic
hemodynamic instability
Minimally invasive
volume; EVLW, extravascular lung water; IJ, internal jugular; RMS,
r-Base
System
b
P P
c p e t S
a s p d w a w a o i r c
g H s v n o
h
P N
t
4 PAUL E. MARIK
edema.49,52-54 The monitor also measures SVV/PVV, which has een shown to be predictive of fluid responsiveness.55 In a
randomized controlled trial, Mutoh et al56 showed an improved linical outcome for patients with subarachnoid hemorrhage andomized to a PiCCO-based hemodynamic algorithm as ompared with the “standard of care,” which used a PAC lgorithm. Additional studies are required to evaluate the clin- cal benefit of this technology.
ulse Contour Requiring Patient Demographic and hysical Characteristics and No Calibration
The FloTrac system consists of the FloTrac sensor and orresponding Vigileo monitor. The system is operator inde- endent, needs no external calibration, and requires a periph- ral arterial catheter only. The basic principle of the system is he linear relation between the pulse pressure and the SV. The V is estimated using the following equation43: SV SDAP
. The arterial pressure waveform is sampled each 20 seconds t 100 Hz, which results in 2,000 data points. SDAP is the tandard deviation of these data points and reflects the pulse ressure. The factor represents the conversion factor that epends on arterial compliance, the mean arterial pressure, and aveform characteristics. The patient’s vascular compliance is
ssessed using biometric values (ie, sex, age, height, and eight) according to the method described by Langewouters et
l.57 Waveform characteristics assessed are skewness (degree f asymmetry) and kurtosis (degree of peakedness) of the ndividual arterial pressure curve. Skewness and kurtosis rep- esent changes in the arterial waveform, which should reflect hanges in vascular tone. The factor is recalculated every
minute and enables calculation of the SV without external calibration.
Because the system is operator independent, easy to use, needs no external calibration, and only requires a peripheral arterial catheter (usually the radial artery), the FloTrac system has found popular appeal and has been studied widely, partic- ularly in the setting of cardiac surgery. To date, the accuracy of FloTrac has been evaluated in 45 studies58-102; these are sum- marized in Table 2. Studies evaluating the first-generation FloTrac showed poor agreement compared with intermittent thermodilution, which is the gold standard. Second-generation devices were purported to be more reliable; however, their accuracy remained clinically unacceptable. Furthermore, in pa- tients with low SVR (eg, sepsis or liver failure), measurements were unreliable, with the bias being correlated with the SVR.65,95,99,101 The data from Table 2 show that the percentage error is lower in cardiac patients as compared with other cohorts (37% 11% v 47% 11%, p 0.01). The third- eneration software claims to have overcome these problems. owever, 6 recent validation studies evaluating this latest ver-
ion do not show improved accuracy in comparison with older ersions.97-102 More problematic is the fact that the system does ot track changes in the SV accurately after a volume challenge r after the use of vasopressors.65,79,85,94,95,100,101,103 These lim-
itations significantly restrict the clinical use of this device. The SVV may be useful in intraoperative fluid optimization in select noncardiac surgical patients.104 However, in a co-
ort of medical patients, it was reported that the SVV was d
oorly predictive of volume responsiveness.105 Takala et l106 randomized 388 hemodynamically unstable patients to oninvasive monitoring with the FloTrac system for 24 ours or usual care (the control group). The main outcome easure was the proportion of patients achieving hemody-
amic stability within 6 hours of starting the study. Surpris- ngly, the time to reach the predefined resuscitation goals as longer in the FloTrac group, with worse clinical out-
omes in these patients.
ulse Contour Requiring No Patient Data and o Calibration
The MostCare system uses the pressure recording analytic ethod (PRAM) to determine the SV.107 PRAM measures the
area under the curve of the arterial waveform. No external calibration or pre-estimated data are required. The morphology of the arterial waveform is analyzed to determine an internal calibration. This system uses high-time resolution by sampling the signal at 1,000 Hz, and it analyzes the whole cardiac…
Related Documents
![NONINVASIVE ECG IMAGING [ ECGI ] OF CARDIAC ARRHYTHMIAS · electrocardiographic measurements, noninvasively • Noninvasive imaging is a corner stone of the practice of modern medicine](https://static.cupdf.com/doc/110x72/5f022a177e708231d402e402/noninvasive-ecg-imaging-ecgi-of-cardiac-arrhythmias-electrocardiographic-measurements.jpg)
